INDEX

Explanations

proper nouns or entities

oai_token-act-pair · gpt-3.5-turbo

sequential elements or patterns, particularly focusing on alphabetical characters and their arrangements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 guiActiveUn

-0.79

rhy

-0.70

igs

-0.68

Reviewed

-0.68

poke

-0.67

epad

-0.66

uner

-0.63

wcsstore

-0.62

idious

-0.62

 transact

-0.60

POSITIVE LOGITS

 Agency

0.72

 Virgin

0.68

 Fields

0.68

 Hill

0.67

 Retrieved

0.67

 Cellular

0.65

 Border

0.64

 Municip

0.63

 Geological

0.63

 Borders

0.62

Activations Density 0.083%

proper nouns or entities

sequential elements or patterns, particularly focusing on alphabetical characters and their arrangements

No Comments

No Known Activations